Author: Claude (AI), from PubMed abstracts · Reviewed by: Dr Mina Abdelmseih · Reviewed: 2026-11-15 · Review due: 2027-11-15

Educational content only. Not personalised medical advice. Clinical decisions remain the responsibility of the treating clinician.

Choroidal naevus versus small melanoma

Choroidal naevus is a common benign melanocytic tumour, usually found incidentally, with a small risk (<1%) of malignant transformation [S1]. The clinical challenge is to separate it from small choroidal melanoma, with which it overlaps as "indeterminate" lesions [S7]. Imaging-based risk factors identify the minority that need specialist review [S2].

Key points - Postequatorial choroidal naevi are present in approximately 5% of US adults aged 40 years or more [S1]. - Risk factors for transformation are summarised by the mnemonic TFSOM-DIM: thickness >2 mm, subretinal fluid, symptoms of vision loss, orange pigment, ultrasonographic hollowness and diameter >5 mm [S2]. - Five-year risk of growth into melanoma rose from 1.1% with no risk factor to 55% with five risk factors in a cohort of 3806 naevi [S2]. - Lesions with three or more risk factors should be evaluated at an experienced ocular oncology centre [S1]. - Most indeterminate lesions that grow do so in the first 2 years of observation [S8].

Evidence last checked: 2026-10-17 (PubMed abstracts).

Red flags

• Multiple risk factors. Three or more risk factors imply more than 50% chance of transformation to melanoma within 5 years in one review [S1]; in the imaging cohort, 5-year growth was 34% with three factors and 51% with four [S2].
• Documented growth. Among 8606 indeterminate tumours followed, 478 (5.6%) grew over 5 years; early growth may represent melanoma in evolution, and late growth after stability may represent malignant transformation [S8].
• Giant naevus. Basal dimension of 10 mm or more carries greater risk of transformation [S1].
• Symptoms. Flashes, floaters or blurred vision are listed risk factors [S1].

Recognise it

• A naevus is generally well circumscribed and pigmented; the mean basal dimension in population studies is approximately 1.25 mm [S1].
• Older risk factors also include margin less than 3 mm from the optic disc, and absence of halo and of drusen [S1].
• Many benign pigmented fundus lesions mimic naevus or melanoma, including congenital hypertrophy of the retinal pigment epithelium, RPE hyperplasia, melanocytoma and others [S5].

Confirm it

• Ultrasonography: thickness >2 mm and acoustic hollowness are risk factors [S2].
• OCT: detects subretinal fluid [S2]; enhanced-depth and swept-source OCT can show subretinal fluid, RPE disruption and dome-shaped elevation, and OCT angiography helps separate melanoma from naevi [S9]. OCT angiography showed naevi with minimal overlying macular microvascular change compared with melanoma [S1].
• Fundus autofluorescence: identifies orange pigment (lipofuscin) [S2] [S4].
• Fundus photography: measures diameter >5 mm [S2] and documents change over time; a white paper describes multimodal imaging and a possible future role for artificial intelligence in detecting transformation [S3].
• Biopsy: can be used for diagnosis of indeterminate lesions or prognosis in small melanoma [S7].

Manage it

Before acting on this section, check the national and local guidelines, product information and regulatory approvals that apply where you practise. See the medical disclaimer.

Tier A (primary care / no specialist)

• A pigmented fundus lesion seen on fundoscopy or retinal photography should be photographed if possible and assessed by an eye clinic; dilated examination and imaging are needed to grade risk [S2] [S5].
• Ask about flashes, floaters or blurred vision [S1].

Tier B (general eye clinic)

• Record diameter and margin-to-disc distance with fundus photography and look for orange pigment, fluid, drusen and halo [S1] [S2].
• Use ultrasonography and OCT where available to measure thickness and detect subretinal fluid [S2].
• Observation is reasonable for asymptomatic benign lesions [S5].
• Refer lesions with three or more risk factors, documented growth or symptoms (see Refer when) [S1] [S8].

Tier C (specialist centre)

• Indeterminate lesions can be closely monitored for growth with multimodal imaging [S7]. Treating an indeterminate lesion is weighed against visual morbidity from local treatment versus rising metastatic risk with size [S7].
• Small choroidal melanoma treated with plaque radiotherapy had tumour recurrence of 7% and 11% at 5 and 10 years and loss of 3 or more Snellen lines in 39% and 49% [S2].
• Delay matters less for small melanomas than larger ones: modelling estimates a 0.3 percentage point increase in 10-year metastatic death per month untreated for small melanomas, against 1.8 and 4.0 for medium and large tumours [S6]. Small melanomas had a mean volume doubling time of 717 days, growing naevi 6392 days [S6].

Refer when

Pitfalls

• Relying on fundoscopy alone: several diagnostic features can be missed without multimodal imaging [S5].
• Assuming a lesion stable for years cannot change: late growth may mean malignant transformation [S8].
• Treating every indeterminate lesion: treatment carries visual morbidity [S7].
• Ignoring symptoms such as flashes or blurred vision [S1].

Evidence notes

• Thin evidence: apart from one meta-analysis [S6], the sources are narrative reviews; risk factor figures come from single-centre cohorts summarised in reviews [S1] [S2].
• The two mnemonics differ: the older one includes margin near disc, halo absence and drusen absence [S1]; the imaging-based one adds diameter >5 mm and drops those factors [S2].
• Prevalence figures are from US adults and naevi are found predominantly in White people, so they may not apply elsewhere [S1].
• Not covered by these sources: evidence-based follow-up intervals for low-risk naevi, prevalence in African and Asian populations, and validated referral timings.

Related articles

• Uveal melanoma: diagnosis and primary treatment
• Choroidal metastases

References

See the numbered source list below. Each marker [S#] links to its PubMed and DOI record.